CN110532545A - A kind of data information abstracting method based on complex neural network modeling - Google Patents
A kind of data information abstracting method based on complex neural network modeling Download PDFInfo
- Publication number
- CN110532545A CN110532545A CN201910665705.8A CN201910665705A CN110532545A CN 110532545 A CN110532545 A CN 110532545A CN 201910665705 A CN201910665705 A CN 201910665705A CN 110532545 A CN110532545 A CN 110532545A
- Authority
- CN
- China
- Prior art keywords
- neural network
- module
- complex neural
- data information
- output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 28
- 238000003062 neural network model Methods 0.000 claims abstract description 32
- 238000000605 extraction Methods 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 14
- 238000005457 optimization Methods 0.000 claims abstract description 8
- 210000002569 neuron Anatomy 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 12
- 150000001875 compounds Chemical class 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000002834 transmittance Methods 0.000 claims description 3
- 230000000644 propagated effect Effects 0.000 claims description 2
- 239000000284 extract Substances 0.000 abstract description 3
- 239000010410 layer Substances 0.000 description 17
- 230000006870 function Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 230000007423 decrease Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 201000003373 familial cold autoinflammatory syndrome 3 Diseases 0.000 description 3
- 239000011229 interlayer Substances 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000016776 visual perception Effects 0.000 description 3
- 238000011478 gradient descent method Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006854 communication Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
A kind of data information abstracting method based on complex neural network modeling, method and step includes: to establish training module and complex neural network model；Complex neural network model is trained using training module；Data information is collected, database module is constructed；Information extraction is carried out to database module using the complex neural network model of optimization.The present invention extracts information on the basis of the complex neural network model of BP and CNN, the cooperation, complementation of the two, so that information extraction accuracy is high, speed is fast, meets currently the needs of to information extraction.
Description
Technical field
The present invention relates to data informations to extract field more particularly to a kind of data information based on complex neural network modeling
Abstracting method.
Background technique
The research hotspot that artificial intelligence field rises since artificial neural network is the 1980s.It is from information
Reason angle is abstracted human brain neuroid, establishes certain naive model, different nets is formed by different connection types
Network.Neural network or neural network are also often directly referred to as in engineering and academia.Neural network is a kind of operational model, by
Composition is coupled to each other between a large amount of node (or neuron).A kind of each specific output function of node on behalf referred to as swashs
Encourage function.Connection between every two node all represents a weighted value for passing through the connection signal, referred to as weight, this phase
When in the memory of artificial neural network.The output of network then according to the connection type of network, the difference of weighted value and excitation function and
It is different.Recently during the last ten years, the research work of artificial neural network deepens continuously, and very big progress is had been achieved for, in mould
The fields such as formula identification, intelligent robot, automatic control, predictive estimation, biology, medicine, economy have successfully solved many existing
For the insoluble practical problem of computer, good intelligent characteristic is shown, the type of artificial neural network is also gradually more
Sample, wherein BP and CNN are most widely used.
Wherein BP is a kind of Multi-layered Feedforward Networks by error back propagation training, its basic thought is gradient decline
Method, using gradient search technology, to make the real output value of network and the error mean square difference minimum of desired output；CNN
Be it is a kind of copy the visual perception mechanism construction of biology comprising convolutional calculation and with the feedforward neural network of depth structure, can be with
Exercise supervision study and unsupervised learning, and the sparsity that the convolution kernel parameter sharing in hidden layer is connected with interlayer makes CNN
Feature can be revealed with lesser calculation amount plaid matching.But when two kinds of neural networks are used alone, still there is certain limitation,
Such as when extracting to data information, individually a kind of Accuracy and high efficiency of neural network model is no longer satisfied people
Demand.
To solve the above problems, proposing a kind of data information extraction side based on complex neural network modeling in the application
Method.
Summary of the invention
(1) goal of the invention
To solve technical problem present in background technique, the present invention proposes a kind of number based on complex neural network modeling
According to information extraction method, the present invention extracts information, the two on the basis of the complex neural network model of BP and CNN
Cooperation, complementation so that information extraction accuracy is high, speed is fast, meets currently the needs of to information extraction.
(2) technical solution
To solve the above problems, the present invention provides a kind of data information extraction sides based on complex neural network modeling
Method includes the following steps:
S1, training module and complex neural network model are established, wherein complex neural network model is compound by CNN and BP
It constitutes；
S2, complex neural network model is trained using training module, the complex neural network model optimized；
The complex neural network model wherein optimized includes input module, hidden module and output module；
S3, data information is collected, constructs database module；
S4, information extraction is carried out to database module using the complex neural network model of optimization；Extraction process are as follows:
A1, command signal is assigned to the complex neural network model of optimization by input module；
A2, signal are delivered to hidden module, and transmit step by step between neuron along positive direction；
If a3, obtaining matched data information in transmittance process, output module recalls data information, power cut-off
Journey；
If a4, signal are transferred to the bottom end of neural network, matched data information is not obtained yet, then is transferred to backpropagation, is believed
Number along original route return；In return course, the weight and threshold value of neurons at different levels are modified, reduces signal errors；
A5, signal modify the weight of neurons at different levels constantly along both forward and reverse directions back and forth between neurons at different levels simultaneously
And threshold value, until obtaining matched data information, output module recalls data information, power cut-off process.
Preferably, in S2, the forward direction including signal is trained to complex neural network model using training module and is passed
Broadcast two processes of backpropagation with error.
Preferably, it is carried out when calculating error output by from the direction for being input to output, and adjusts weight and threshold value then from defeated
The direction for arriving input out carries out；When forward-propagating, input module input signal acts on output node by hidden module, passes through
Nonlinear transformation generates output signal by output module and is transferred to the anti-of error if reality output is not consistent with desired output
To communication process；Error-duration model is by hidden module by output error to the layer-by-layer anti-pass of input module, and by error distribution to each
All neurons of layer, the error signal obtained from each layer neuron adjust the weight and threshold value of each neuron.By adjusting input
The linking intensity and threshold value of Module nodes and hidden Module nodes decline error along gradient direction, train by repetition learning,
Determine network parameter (weight and threshold value) corresponding with minimal error, training stops stopping.
Preferably, convolutional layer is provided in hidden module.
Preferably, for carrying out the convolutional layer of feature extraction to input data, internal includes multiple convolution kernels；Convolution kernel
At work, it does matrix element multiplication to input feature vector to sum and be superimposed departure, calculating function is
Wherein b is departure,
ZlAnd Zl+1Indicate that l+1 layers of convolution is output and input, also referred to as characteristic pattern, Ll+1For Zl+1Size, it is assumed here that feature
Figure length and width are identical, and the pixel of Z (i, j) character pair figure, K is characterized the port number of figure, f, s0It is convolution layer parameter with p, it is corresponding
Convolution kernel size, convolution step-length and the filling number of plies.
Above-mentioned technical proposal of the invention has following beneficial technical effect:
The present invention extracts information on the basis of the complex neural network model of BP and CNN, and wherein BP is a kind of
By the Multi-layered Feedforward Networks of error back propagation training, its basic thought is gradient descent method, using gradient search technology, with
Phase makes the real output value of network and the error mean square difference minimum of desired output；CNN is a kind of comprising convolutional calculation and tool
There is the feedforward neural network of depth structure, copy the visual perception mechanism construction of biology, can exercise supervision study and non-supervisory
It practises, the sparsity that the convolution kernel parameter sharing in hidden layer is connected with interlayer enables CNN with lesser calculation amount plaid matching
Reveal feature；The cooperation, complementation of the two, so that information extraction accuracy is high, speed is fast, meets currently to the need of information extraction
It asks.
Detailed description of the invention
Fig. 1 is a kind of flow chart of data information abstracting method based on complex neural network modeling proposed by the present invention.
Specific embodiment
In order to make the objectives, technical solutions and advantages of the present invention clearer, With reference to embodiment and join
According to attached drawing, the present invention is described in more detail.It should be understood that these descriptions are merely illustrative, and it is not intended to limit this hair
Bright range.In addition, in the following description, descriptions of well-known structures and technologies are omitted, to avoid this is unnecessarily obscured
The concept of invention.
As shown in Figure 1, a kind of data information abstracting method based on complex neural network modeling proposed by the present invention, including
Following steps:
S1, training module and complex neural network model are established, wherein complex neural network model is compound by CNN and BP
It constitutes；
S2, complex neural network model is trained using training module, the complex neural network model optimized；
The complex neural network model wherein optimized includes input module, hidden module and output module；
S3, data information is collected, constructs database module；
S4, information extraction is carried out to database module using the complex neural network model of optimization；Extraction process are as follows:
A1, command signal is assigned to the complex neural network model of optimization by input module；
A2, signal are delivered to hidden module, and transmit step by step between neuron along positive direction；
If a3, obtaining matched data information in transmittance process, output module recalls data information, power cut-off
Journey；
If a4, signal are transferred to the bottom end of neural network, matched data information is not obtained yet, then is transferred to backpropagation, is believed
Number along original route return；In return course, the weight and threshold value of neurons at different levels are modified, reduces signal errors；
A5, signal modify the weight of neurons at different levels constantly along both forward and reverse directions back and forth between neurons at different levels simultaneously
And threshold value, until obtaining matched data information, output module recalls data information, power cut-off process.
In an alternative embodiment, in S2, packet is trained to complex neural network model using training module
Include two processes of backpropagation of the propagated forward and error of signal.
In an alternative embodiment, it is carried out when calculating error output by from the direction for being input to output, and adjusts power
Value and threshold value are then carried out from the direction for being output to input；When forward-propagating, input module input signal is acted on by hidden module
Output node generates output signal by output module by nonlinear transformation, if reality output is not consistent with desired output,
Then it is transferred to the back-propagation process of error；Error-duration model is by hidden module by output error to the layer-by-layer anti-pass of input module, and
Give error distribution to all neurons of each layer, the error signal obtained from each layer neuron adjusts the weight and threshold of each neuron
Value.By adjusting the linking intensity and threshold value of input module node and hidden Module nodes, decline error along gradient direction, warp
Repetition learning training is crossed, determines network parameter (weight and threshold value) corresponding with minimal error, training stops stopping.
In an alternative embodiment, convolutional layer is provided in hidden module.
In an alternative embodiment, internal comprising more for carrying out the convolutional layer of feature extraction to input data
A convolution kernel；Convolution kernel at work, does matrix element multiplication to input feature vector and sums and be superimposed departure, calculating function is
Wherein b is departure,
ZlAnd Zl+1Indicate that l+1 layers of convolution is output and input, also referred to as characteristic pattern, Ll+1For Zl+1Size, it is assumed here that feature
Figure length and width are identical, and the pixel of Z (i, j) character pair figure, K is characterized the port number of figure, f, s0It is convolution layer parameter with p, it is corresponding
Convolution kernel size, convolution step-length and the filling number of plies.
In the present invention, on the basis of the complex neural network model of BP and CNN, information is extracted, wherein BP is
A kind of Multi-layered Feedforward Networks by error back propagation training, its basic thought is gradient descent method, utilizes gradient search skill
Art, to make the real output value of network and the error mean square difference minimum of desired output；CNN is a kind of comprising convolution meter
Calculate and with depth structure feedforward neural network, copy biology visual perception mechanism construction, can exercise supervision study and it is non-
Supervised learning, the sparsity that the convolution kernel parameter sharing in hidden layer is connected with interlayer enable CNN with lesser calculating
Amount plaid matching reveals feature；The cooperation, complementation of the two, so that information extraction accuracy is high, speed is fast, meets and currently takes out to information
The demand taken.
It should be understood that above-mentioned specific embodiment of the invention is used only for exemplary illustration or explains of the invention
Principle, but not to limit the present invention.Therefore, that is done without departing from the spirit and scope of the present invention is any
Modification, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.In addition, appended claims purport of the present invention
Covering the whole variations fallen into attached claim scope and boundary or this range and the equivalent form on boundary and is repairing
Change example.
Claims (5)
1. a kind of data information abstracting method based on complex neural network modeling, which comprises the steps of:
S1, training module and complex neural network model are established, wherein complex neural network model is constituted by CNN and BP are compound；
S2, complex neural network model is trained using training module, the complex neural network model optimized；Wherein
The complex neural network model of optimization includes input module, hidden module and output module；
S3, data information is collected, constructs database module；
S4, information extraction is carried out to database module using the complex neural network model of optimization；Extraction process are as follows:
A1, command signal is assigned to the complex neural network model of optimization by input module；
A2, signal are delivered to hidden module, and transmit step by step between neuron along positive direction；
If a3, obtaining matched data information in transmittance process, output module recalls data information, power cut-off process；
If a4, signal are transferred to the bottom end of neural network, matched data information is not obtained yet, then is transferred to backpropagation, signal edge
Original route returns；In return course, the weight and threshold value of neurons at different levels are modified, reduces signal errors；
A5, signal modify the weight and threshold of neurons at different levels constantly along both forward and reverse directions back and forth between neurons at different levels simultaneously
Value, until obtaining matched data information, output module recalls data information, power cut-off process.
2. a kind of data information abstracting method based on complex neural network modeling according to claim 1, feature exist
In being trained the anti-of propagated forward including signal and error to complex neural network model using training module in S2
To two processes of propagation.
3. a kind of data information abstracting method based on complex neural network modeling according to claim 2, feature exist
In, it is carried out when calculating error output by from the direction for being input to output, and weight and threshold value are adjusted then from the side for being output to input
To progress；When forward-propagating, input module input signal acts on output node by hidden module, by nonlinear transformation, leads to
It crosses output module generation output signal and is transferred to the back-propagation process of error if reality output is not consistent with desired output；Accidentally
Poor anti-pass is by hidden module by output error to the layer-by-layer anti-pass of input module, and gives error distribution to all neurons of each layer,
The error signal obtained from each layer neuron, adjusts the weight and threshold value of each neuron.
4. a kind of data information abstracting method based on complex neural network modeling according to claim 1, feature exist
In being provided with convolutional layer in hidden module.
5. a kind of data information abstracting method based on complex neural network modeling according to claim 4, feature exist
In for carrying out the convolutional layer of feature extraction to input data, internal includes multiple convolution kernels；Convolution kernel is at work, right
Input feature vector does matrix element multiplication and sums and be superimposed departure, calculates function and is
Wherein b is departure, ZlWith
Zl+1Indicate that l+1 layers of convolution is output and input, also referred to as characteristic pattern, Ll+1For Zl+1Size, it is assumed here that characteristic pattern is long
Width is identical, and the pixel of Z (i, j) character pair figure, K is characterized the port number of figure, f, s0It is convolution layer parameter with p, corresponding convolution
Core size, convolution step-length and the filling number of plies.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910665705.8A CN110532545A (en) | 2019-07-23 | 2019-07-23 | A kind of data information abstracting method based on complex neural network modeling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910665705.8A CN110532545A (en) | 2019-07-23 | 2019-07-23 | A kind of data information abstracting method based on complex neural network modeling |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110532545A true CN110532545A (en) | 2019-12-03 |
Family
ID=68661714
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910665705.8A Pending CN110532545A (en) | 2019-07-23 | 2019-07-23 | A kind of data information abstracting method based on complex neural network modeling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110532545A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111831657A (en) * | 2020-06-28 | 2020-10-27 | 苏州市测绘院有限责任公司 | Leveling data real-time processing method |
| CN111881034A (en) * | 2020-07-23 | 2020-11-03 | 深圳慕智科技有限公司 | Confrontation sample generation method based on distance |
| CN113177997A (en) * | 2021-04-25 | 2021-07-27 | 上海雷鸣文化传播有限公司 | Method for realizing ink and wash effect rendering based on neural network graphic algorithm |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7069256B1 (en) * | 2002-05-23 | 2006-06-27 | Oracle International Corporation | Neural network module for data mining |
| CN106845478A (en) * | 2016-12-30 | 2017-06-13 | 同观科技（深圳）有限公司 | The secondary licence plate recognition method and device of a kind of character confidence level |
| CN108460742A (en) * | 2018-03-14 | 2018-08-28 | 日照职业技术学院 | A kind of image recovery method based on BP neural network |
-
2019
- 2019-07-23 CN CN201910665705.8A patent/CN110532545A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7069256B1 (en) * | 2002-05-23 | 2006-06-27 | Oracle International Corporation | Neural network module for data mining |
| CN106845478A (en) * | 2016-12-30 | 2017-06-13 | 同观科技（深圳）有限公司 | The secondary licence plate recognition method and device of a kind of character confidence level |
| CN108460742A (en) * | 2018-03-14 | 2018-08-28 | 日照职业技术学院 | A kind of image recovery method based on BP neural network |
Non-Patent Citations (1)
| Title |
|---|
| 姚红革等: "基于小波分析和BP神经网络的图像特征提取", 《西安工业大学学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111831657A (en) * | 2020-06-28 | 2020-10-27 | 苏州市测绘院有限责任公司 | Leveling data real-time processing method |
| CN111881034A (en) * | 2020-07-23 | 2020-11-03 | 深圳慕智科技有限公司 | Confrontation sample generation method based on distance |
| CN113177997A (en) * | 2021-04-25 | 2021-07-27 | 上海雷鸣文化传播有限公司 | Method for realizing ink and wash effect rendering based on neural network graphic algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Prediction for tourism flow based on LSTM neural network | |
| CN107247989A (en) | A kind of neural network training method and device | |
| CN104978580B (en) | A kind of insulator recognition methods for unmanned plane inspection transmission line of electricity | |
| CN110532545A (en) | A kind of data information abstracting method based on complex neural network modeling | |
| CN105160400B (en) | The method of lifting convolutional neural networks generalization ability based on L21 norms | |
| CN110060475A (en) | A kind of multi-intersection signal lamp cooperative control method based on deeply study | |
| CN109241291A (en) | Knowledge mapping optimal path inquiry system and method based on deeply study | |
| CN110223517A (en) | Short-term traffic flow forecast method based on temporal correlation | |
| CN106558058A (en) | Parted pattern training method, lane segmentation method, control method for vehicle and device | |
| CN110334705A (en) | A kind of Language Identification of the scene text image of the global and local information of combination | |
| CN109492556A (en) | Synthetic aperture radar target identification method towards the study of small sample residual error | |
| CN104217214A (en) | Configurable convolutional neural network based red green blue-distance (RGB-D) figure behavior identification method | |
| CN109063569A (en) | A kind of semantic class change detecting method based on remote sensing image | |
| CN108021947A (en) | A kind of layering extreme learning machine target identification method of view-based access control model | |
| CN109558902A (en) | A kind of fast target detection method | |
| CN113205595B (en) | Construction method and application of 3D human body posture estimation model | |
| CN108710906A (en) | Real-time point cloud model sorting technique based on lightweight network LightPointNet | |
| CN109241995A (en) | A kind of image-recognizing method based on modified ArcFace loss function | |
| CN106650918B (en) | The method and apparatus for constructing system model | |
| Qu et al. | Visual cross-image fusion using deep neural networks for image edge detection | |
| Pérez et al. | Physics-Informed Long-Short Term Memory Neural Network Performance on Holloman High-Speed Test Track Sled Study | |
| CN106896724A (en) | For the tracking system and tracking of sun tracker | |
| CN110503148A (en) | A kind of point cloud object identifying method with scale invariability | |
| CN110188621A (en) | A kind of three-dimensional face expression recognition methods based on SSF-IL-CNN | |
| CN109543748A (en) | A kind of fruit recognition methods based on multi-model fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220913 Address after: 361000 units 1702 and 1703, No. 59, Chengyi North Street, phase III, software park, Xiamen, Fujian Applicant after: XIAMEN USEEAR INFORMATION TECHNOLOGY Co.,Ltd. Address before: Unit 1701, unit 1704, No. 59, Chengyi North Street, phase III, software park, Xiamen City, Fujian Province, 361000 Applicant before: FUJIAN QIDIAN SPACE-TIME DIGITAL TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191203 |
|
| RJ01 | Rejection of invention patent application after publication |